An AI agent that applies to roles like this on employers' own sites. 25 applications free, no card needed.
Posted over 15 days ago, so the role may have closed. Confirm on the employer's page when you apply, or see similar jobs.
Job Description
Role & Responsibilities
- Design, write, and optimize high-performance CUDA kernels targeting NVIDIA Ampere, Hopper, and upcoming architectures.
- Profile and debug GPU memory bandwidth, occupancy, and instruction-level bottlenecks using Nsight Compute, Visual Profiler, and custom instrumentation.
- Collaborate with ML and HPC teams to translate algorithmic requirements into efficient, scalable, and portable GPU code.
- Implement fused operations, shared memory optimizations, and warp-level primitives to maximize throughput under latency and memory constraints.
- Contribute to kernel library development and maintain reusable, well-documented, and performance-tested modules.
- Stay ahead of evolving CUDA toolchains, compiler flags, and architecture-specific features to continuously extract performance wins.
Must-Have
- CUDA C++
- Nsight Compute
- NVIDIA GPU Architecture (Ampere/Hopper)
- GPU Memory Hierarchy Optimization
- Warp-Level Primitives
- Profiling & Performance Tuning
- nvcc Compiler
- PTX / SASS Assembly (Basic Understanding)
- Experience with Tensor Cores and FP8/FP16 kernels
- familiarity with cuDNN, cuBLAS, or CUTLASS
- Contributions to open-source GPU compute projects
Skills: cuda,graphs,architecture,layout,design,contribute,fusion,kernel,c,computing
Skills
CIf a job posting appears fraudulent, asks for payment, contains misleading information, or violates our guidelines, please report it immediately. Our team will review it promptly, Jobaaj does not charge any fee from the applicants.
Similar Jobs
View all
New IT / Software Development & Related roles in Bengaluru / Bangalore.
One email a day. Unsubscribe any time.
